This is a medical malpractice case. Steven and Christine Horvath, individually and as next friends of their son, Steven Hor-vath Jr., brought suit against Baylor University Medical Center. They alleged that the hospital was negligent in failing to cause their child to be tested for phenylke-tonuria (PKU). The jury found that hospital negligence, if any, was not a proximate cause of the child’s damages. We affirm.
Steven Horvath, Jr. is afflicted with phe-nylketonuria. PKU is a hereditary absence of the enzyme that breaks down the amino acid, phenylalanine. Without the enzyme, phenylalanine builds up to toxic levels and mental retardation results. In 1962, a simple blood test was developed to screen newborns for PKU. Once detected, PKU in*868fants can be treated with a special low protein diet. Fcr treated patients, the outlook is bright; most retain their normal intelligence. It follows that early detection and treatment are of the utmost importance.
Here, the plaintiffs’ child was not so fortunate. He was born at defendant hospital on December 8, 1972. The following day his pediatrician examined him. The child appeared normal and healthy and the doctor gave orders that the infant be started on Similac formula. The feeding began at 1:30 p.m. Mother and baby were discharged at 10:30 a.m. on the next day, December 10. Hospital policy required that infants be routinely screened for PKU on the day of discharge, but the evidence reflected a hospital practice to omit the test when the child was discharged less than twenty-four hours after the first milk feeding — apparently on the premise that a test given that soon would not be reliable. Mother and child left the hospital some twenty-one hours after the first feeding and the infant was not tested. The pediatrician also failed to notify the parents that they should secure a PKU test after discharge. By the time that the child’s condition was detected in his eighth year, the damage was irreversible.
This action was brought against both the pediatrician and defendant hospital, but it was voluntarily dismissed against the doctor before trial commenced. At the trial, plaintiffs urged that the hospital was negligent (1) in discharging the child without testing him; (2) in failing to inform the pediatrician that the child was untested; and (3) in failing to establish procedures that ensured that all infants were screened for PKU.
In response to an inquiry asking, “Was any negligence of [the hospital] a proximate cause of the occurrence in question,” the jury answered “No.” Plaintiffs challenge this conclusion as being so much against the great weight and preponderance of the' evidence as to be manifestly unjust. In essence, they argue that the evidence is factually insufficient to support the jury’s findings.
In reviewing a factual insufficiency point of error, the court must examine the entire record including any evidence contrary to the trial court’s judgment. Burnett v. Motyka, 610 S.W.2d 735, 736 (Tex.1980); In re King’s Estate, 150 Tex. 662, 244 S.W.2d 660 (1951). The appellate court may not substitute its judgment for the trier of fact. Where the evidence conflicts, the jury’s decision will be left undisturbed. Walter Baxter Seed Co. v. Rivera, 677 S.W.2d 241 (Tex.App.—Corpus Christi 1984, writ ref’d n.r.e.).
Plaintiffs called several expert witnesses. Dr. Robert Guthrie, developer of the PKU test, testified that it was the hospital’s obligation to test for PKU and that the substantial likelihood that an infant would not see a doctor after discharge from the hospital dictated that all infants be screened regardless of the time from the first milk feed. He testified that, if a sample had been taken upon Stephen’s discharge, it would have revealed an elevated phenylalanine level. The witness stated that a physician should be able to rely on the hospital to cause the test to be made.
Dr. Charles Sweet, who was the Assistant Director of the Bureau of Laboratories of the Texas Department of Health at the time of the child’s birth in 1972, testified that while a twenty-four hour waiting period was advisable, all infants should be tested before leaving the hospital.
Dr. Julian Williams of the Texas Health Science Center testified that the accepted practice in Texas in 1972 was to test all infants before they left the hospital.
Williams’s opinion was confirmed by Dr. Richard Koch, head of the Collaborative Study on Phenylketonuria. He stated that testing for all infants has been a national standard since the late 1960’s. He, too, expressed concern that as many as one-fourth of the children born in the United States do not see a pediatrician after discharge.
Dr. Stephen Moore, a Dallas pediatrician, testified that he relied on the staff at the hospital to perform the PKU tests. On the *869rare occasions when the test had not been performed, the hospital staff notified him.
But the hospital also brought forward expert testimony. Dr. George Branch, a Dallas pediatrician, testified that when this birth occurred in 1972 a PKU test on an infant who had not been on milk feed for twenty-four hours was considered medically useless. He disputed Dr. Moore’s claim that it was the hospital’s practice to notify the physician if a test was not performed. Dr. Branch stated that the hospital’s responsibility for testing ceased upon the patient’s discharge; any necessary follow-up fell within the physician’s province.
Another pediatrician, Dr. Doman Keele, testified that the hospital’s practices were within the accepted medical standards of that time. He stated that PKU testing before twenty-four hours of milk feed was considered inaccurate in 1972. Dr. Keele further testified that a hospital would be justified in assuming that a physician would assume responsibility for the PKU testing of infants not screened at the hospital.
Nurse Frances Varner was in charge of the nursery in 1972. Her testimony essentially confirmed that of Dr. Branch. She stated that doctors were not notified if a test was not performed. It was considered the pediatrician’s responsibility.
Much of the dispute at trial centered around the following statutory provision and its implementation:
The physician attending a newborn child, or the person attending a newborn child that was not attended by a physician, shall cause the child to be subjected to the phenylketonuria test that has been approved by the State Department of Health.
TEX.REV.CIV.STAT.ANN. art. 4447e (Vernon 1976), amended by Act of Aug. 27, 1979, ch. 269 §§ 1 2, 4, 5, 1979. Tex.Gen. Laws p. 578; Act of Aug. 29, 1983, ch. 907, § 1, 1983 Tex.Gen.Laws 5035. In connection with the act, the Texas Department of Health disseminated a letter to doctors and hospitals. The letter advised, “A test should be made on a blood specimen collected from each newborn infant as late as possible prior to discharge from the hospital and no sooner than 24 hours after the onset of milk feeding.”
Dr. Charles Sweet and Donald Tausch, two Department of Health employees, testified that the letter’s intent was that all infants should be tested, regardless of discharge time. The twenty-four hour recommendation merely represented the ideal procedure. Sweet acknowledged, however that he had learned that actual hospital practice varied. Some were not testing until up to seventy-two hours after feeding. A form letter from the Department’s Bureau of Laboratories advised persons submitting samples to collect specimens no sooner than forty-eight hours after the first feeding.
From our examination of the record, we believe a jury could have legitimately concluded that the hospital was not negligent or that its negligence was not a proximate cause of the occurrence. Under the broad submission used, we cannot say which of these conclusions was reached. The jury, taking into account the directives of the Health Department, the doubt concerning the efficacy of early testing, and the differing practices of hospitals, could conclude that the hospital did not fall below the accepted standard of care. Similarly, the testimony and the statute in force at the time might well have convinced the jury that the attending physician was primarily responsible for assuring that a PKU test was done and that the pediatrician’s omission to follow up on the matter at post discharge visits in his office broke the chain of proximate causation.
In substance, plaintiffs are asking the court to conclude that the impressive qualifications of their experts outweigh the testimony of defendant’s experts. This, we cannot do. The jury, as exclusive judge of the facts, had the sole responsibility to evaluate the credibility of witnesses and determine the weight to be given to their testimony. First City Bank of Farmer’s Branch v. Guex, 659 S.W.2d 734 (Tex.App.-Dallas 1983) aff'd, 677 S.W.2d 25 (Tex.1984). If the jury system is to retain its *870vitality, the appellate courts must resist usurping the jury’s role. Only verdicts so against the great weight and preponderance of the evidence as to be manifestly unjust warrant reversal. The conflicting testimony presented the jurors with a difficult question. We cannot say that their answer was manifestly unjust.
Plaintiffs’ third and fourth points of error complain of the trial court’s refusal to admit a 1971 United States Department of Health, Education and Welfare publication containing recommended guidelines for PKU screening programs.
Rule 803(8)(C) of the Texas Rules of Evidence permits the introduction of properly authenticated public records, statements, and data compilations. The records are admissible if they fall under one of three categories: (1) description of the agency’s activities; (2) matters observed pursuant to a duty imposed by law as to which matters there was a duty to report; (3) factual findings resulting from an investigation made pursuant to authority granted by law. The proposed guidelines fall into none of these categories. An instrument containing “recommended guidelines” lacks the indicia of reliability displayed by public records of factual events. The publication was properly excluded as hearsay.
Plaintiffs’ final three points of error relate to the trial court’s definition of ordinary care. The court instructed the jury that ordinary care means “that degree of care which would be used by a hospital of ordinary prudence under the same or similar circumstances in December 1972 in Dallas, Texas.” Plaintiffs argue that the inclusion of the phrase, “in Dallas, Texas,” was erroneous because there was either no evidence or insufficient evidence indicating that the Dallas standard differed from the national one.
We find no merit in the contention. First, if the standards were indeed identical nationwide, the inclusion of “in Dallas, Texas” was harmless because Dallas is a part of the nation. Second, there was evidence concerning statutes in several states and regulations affecting those statutes. Because those statutes could have had some bearing on the ultimate issue, the instruction was proper. Finally, the instruction properly reflected the plaintiff’s legal burden. Elizondo v. Tavarez, 596 S.W.2d 667 (Tex.Civ.App.—Corpus Christi 1980, writ ref’d n.r.e.) (burden on plaintiff to produce probative evidence of relevant practice in community where action arose). We find no reversible error in the instruction.
Affirmed.